×

Rush Limbaugh

For a better experience,
download and use our app!

The Rush Limbaugh Show Main Menu




Listen to it Button

RUSH: Nate Silver is out.  This is going to disquiet those on the left.  Trump has surged. Nate Silver… I’ll just explain what he does.  I know you probably have heard me say this, but we have new people tuning in every day here, folks, and they — in many cases — are hearing things for the first time that you’ve heard a lot so indulge me.  Nate Silver used to be at the New York Times.  He’s an analyst of polls.  He’s not a pollster per se.

He looks at everybody else’s polls and then he analyzes them as supposedly only he and his staff can. And then, after that analysis, he attaches a percentage of likelihood of victory to each candidate.  And when this whole thing got started, Nate Silver factored that Trump had a 3% chance of winning the White House.  Today, it’s 48%.  “The latest vote projection from elections guru Nate Silver has Republican Donald Trump just six electoral votes short of winning and one point away from equaling Hillary Clinton’s popular vote. The newest Five Thirty Eight survey Trump at 264 and Clinton at 272, two more than needed. It’s the closest in recent weeks.”

It’s the closest it’s been in recent weeks.  So Trump — just a couple of months ago, according to Nate Silver — had a 3% chance of winning. He’s up now to 48%, and it is climbing.  Nobody has to send me anything here.  I just want to throw something out.  I didn’t print it ’cause it’s too in the weeds and too esoteric, but it’s an interesting thing.  Somebody… Somebody credible, I don’t even remember who — don’t doubt me on this. They took the same data, and this company went out and did a poll.  They polled the American people.

They polled, I think, something like 800 people.  They gave the data to five different pollsters, and five different pollsters produced five different results with the identical data.  And the story purported to explain how this could happen, as though it was entirely understandable.  How is it understandable?  You have the polling questions, you had the answers, you had the cross tabs, you have everything.  In four of the five, pollster analysis, Trump won by a point or two. And in the other four, Hillary was narrowly ahead.

I mean, it was close; it didn’t… None of the five pollsters showed either candidate running away with it.  But it was the same data.  Five different poll companies were given the identical data, and they come up five different interpretations in what it meant.  I was reading the story and said, “Yeah, well, depends on the questions you ask.”  Well, they didn’t ask any questions.  The questions were asked for them.  The answers were provided for them.  Well, yeah, but then you have…

You know, every poll weights. w-e-i-g-h-t. Every poll weights its questions and appearances and that’s up to the individual pollster.  Really?  So the individual pollsters look at the questions being asked and then they decide, “That’s a good question, or not a good question; I don’t think I should give full weight to that”? Is that how happens? “Basically you’re right, Mr. Limbaugh, yeah.”  So bottom line is: Same data, five different companies, five different results.  Just keep that in mind going forward the next time you see any poll.

Pin It on Pinterest

Share This